One of the research goals related to modern beer production is determining the molecular factors for beer quality and stability. A large fraction of beer is produced on the basis of barley (Hordeum vulgare, L.). It is a monocotyledonous crop plant grown in many parts of the world, not only due to its economic importance as a source of industrial products, such as beer, but also as a source of animal feed. The United States is now one of the leading producers of malting barley, with around 13% of the world crop; Canada, Australia and Europe together account for about 70% of the production (Bios Intern., 2001).
A continuing effort of barley breeders is to develop stable, high-yielding cultivars that are agronomically sound. To accomplish this goal, attempts have included random mutagenesis by chemical treatment or irradiation to modify traits of interest, for example to alter the expression of specific genes that may have deleterious effects on plant growth and crop productivity in general—but also on traits conferring added quality to a product manufactured from the crop. It is well established that sodium azide, NaN3, is a useful chemical to mutagenize barley. Specifically, NaN3-derived mutagenesis has been used to induce genetic changes in barley to generate mutants blocked in the synthesis of anthocyanins and proanthocyanidins (von Wettstein et al., 1977; von Wettstein et al., 1985; Jende-Strid, 1991; Jende-Strid, 1993; Olsen et al., 1993). A second example relates to barley kernels mutagenized with NaN3 to screen for high levels of free phosphate with the aim to identify low-phytate mutants (Rasmussen and Hatzak, 1998); a total of 10 mutants out of 2,000 screened kernels were identified. Although a major drawback in barley genetics has been the inability to specifically study gene function through reverse genetics, forward genetic screens—e.g. following NaN3-induced mutagenesis—continue regarding improvements that relate to nutritional and product quality parameters of barley and malt.
Except in a gross and general fashion, a breeder cannot predict the outcome of new plant lines under development in a conventional plant breeding process. This unpredictability is mainly caused by the lack of control at the cellular level, more specifically at the level of nuclear DNA—the complexity of which is enormous. A number of other factors influence the outcome of a plant breeding process, for example the climate and soil quality at the geographical location of plant propagation. As a result, different barley breeders that use conventional techniques will never develop plants with identical traits. In the conventional breeding process, a most difficult task is the identification of plants that are genetically superior, not only with respect to the trait of interest, but also with respect to physiological issues of relevance for plant growth. The selection process is particularly difficult when other confounding traits mask the trait of interest. When present-day plant breeding procedures include DNA sequence determination of the mutated gene, it is at a late stage of the breeding program—i.e. after mutant characterization, for example as recently described for screening of chemically induced mutations in Arabidopsis and other plants (Colbert et al., 2001).
Thus far, the creation of gene-indexed loss-of-function mutations on a whole-genome scale has been reported for the yeast Saccharomyces cerevisiae (Giaever et al., 2002). For the plant Arabidopsis, 21,700 of the ˜29,454 predicted genes have been inactivated by the insertion of Agrobacterium T-DNA sequences (Alonso et al., 2003).
Until now, it is not unusual that a conventional breeding process from the first mutagenesis or crossing to marketing of plants or seeds takes >10 years. Specifically, it would be excellent to provide the plant breeder with methods to detect mutations in the gene related to the trait of interest. Such improvements would enhance the level of predictability in breeding programs, especially when the selection of mutants is directed toward those having nonsense mutations in the protein-coding part of the gene of interest. In other cases, it may also be preferred with an early identification of DNA mutations, for example to cancel further breeding with lines characterized by promoter mutations in the gene of intererest or where other DNA mutations influence expression—simply because environmental or physiological factors could confer reversion of the trait induced by the mutagen. Accordingly, there is a demand for finding alternative ways of detecting mutations of interest early in the breeding program. This should make the entire breeding process faster and economically of higher interest, thus maximizing the amount of grain produced on the land.
A major proportion of the barley produced comprises malting varieties, the kernels of which are converted to malt through processes of controlled steeping, germination, and drying of the barley. A small proportion of the malt is used as ingredients in the food industry, whereas the majority of the malt is subsequently used as the main ingredient in the production of malt-derived beverages, including, but not limited to, beer and whisky. In the brewhouse, milled malt is subjected to a mashing process comprising a step-wise increase in temperature of a malt-water suspension which confers partial, enzymatic degradation and extraction of, for example, the kernel polymers starch and β-glucan. Following filtration, the aqueous mash is boiled with hops to yield the wort. Said wort is subsequently fermented with yeast, giving the beer product which—upon maturation—is bottled. The wort can also be used for the production of non-fermented malt beverages.
Palatability and flavor stability of a beverage is an important factor of relevance to the composition of barley and malt. This is because natural flavor molecules derived from said barley and malt—or generated by the action of enzymes extracted from said barley and malt—may confer undesirable taste characteristics to the final product (Drost et al., 1990). In this respect, formation of the volatile compound giving a cardboard-like flavor appears to be of particular biochemical as well as economic interest. In 1970, the molecule responsible for cardboard-like flavor was isolated and identified as T2N, a nine-carbon (C9) alkenal (Jamieson and Gheluwe, 1970). Since the taste-threshold level for T2N in humans is extremely low, previously determined to be around 0.7 nM or 0.1 ppb (Meilgaard, 1975), products with even minute levels of the aldehyde are regarded as being aged due to the off-flavor taste of the product. Moreover, liberation of T2N from decomposing T2N adducts during beer storage may cause deterioration of the product (Nyborg et al., 1999).
Radioactive labeling studies with plant tissue established that nonenals are derived from the C18 fatty acid linoleic acid, whereas the hexanals and nonadienals are formed from the C18 fatty acid linolenic acid (Grosch and Schwartz, 1971; Phillips and Galliard, 1978). These and numerous subsequent observations—for example as summarized by Tijet et al. (2001), Noordermeer et al. (2001), and Matsui et al. (2003)—have been interpreted as evidence that T2N is formed by the sequential action of LOX pathway-specific enzymes, with the action of LOX representing an early enzymatic step. Consistent with this notion, Kurodo et al. (2003) found that malt contains a heat-stable enzymatic factor which is necessary for the transformation of the products made by LOX into T2N. The barley kernel contains three LOX enzymes known as LOX-1, LOX-2 and LOX-3 (van Mechelen et al., 1999). While LOX-1 catalyzes the formation of 9-HPODE—a precursor of T2N—from linoleic acid, LOX-2 catalyzes the conversion of linoleic acid to 13-HPODE which is further metabolized to hexanal (FIG. 1B), a C6 aldehyde with a taste threshold level of around 0.4 ppm (Meilgaard, supra). Although the product specificity of LOX-3 remains elusive, the very low expression level of the corresponding gene, as shown by van Mechelen et al. (supra), suggests that its contribution to T2N formation is negligible. Research is ongoing to determine if LOX activity is the sole enzymatic source for the generation of linoleic acid hydroperoxide precursors of relevance for the formation of the T2N-specific off-flavors, or whether the process of fatty acid autooxidation contributes as well. It is notable that C18 hydroperoxides can be further converted by more than seven different families of plant and animal enzymes, with all reactions collectively called the LOX pathway (Feussner and Wasternack, 2002); this pathway is also referred to as the oxylipin pathway. Oxylipins, as their name implies, are oxygenated lipid-derived molecules, which result from the oxygenation of unsaturated fatty acids via the LOX reaction and also include any molecules derived from such oxygenated molecules.
Barley kernels and barley plants having a LOX-1 protein characterized by reduced activity were disclosed in PCT application PCT/IB01/00207 published as WO 02/053721A1 to Douma et al. However, said application does not teach the generation and analysis of barley kernels with inactive LOX-1 enzyme.
Several examples on mutated plants that synthesize low levels of LOX are known. For example, three soybean lines were identified in the early 1980s, each deficient in one of the three LOX enzymes in mature soybean seed:                (i) LOX-1. Although the molecular basis of the LOX-1 null mutation remains uncertain, it correlates with the absence of the corresponding mature mRNA (Hildebrandt and Hymowitz, 1982; Start et al., 1986);        (ii) LOX-2. Transcripts for the mutated gene were detected, and a single base change was observed which replaces a histidine ligand to the active site iron, leading to enzyme instability (Davies and Nielsen, 1986; Wang et al., 1994);        (iii) LOX-3. LOX-3 null mutants exhibited no detectable levels of the corresponding transcript, probably as a consequence of cis-acting elements in the gene promoter (Kitamura et al., 1983; Wang et al., 1995).        
In pea seed, a null-LOX-2 line was found to carry a defect leading to the absence of most LOX-2 protein (Forster et al., 1999). Since this line exhibited a great decrease in the amount of mRNA for LOX-2, it was suggested that the mutation caused a dramatic reduction in mRNA stability.
In rice, immunoblot screening of extracts revealed the presence of two natural cultivars, Daw Dam and CI-115, each lacking one of three LOX enzymes (Ramezanzadeh et al., 1999). It was determined that the amount of hexanal, pentanal, and pentanol in normal rice with all three LOXs was markedly induced during storage, while that in Daw Dam and CI-115 was reduced in the range from 66% to 80%. Despite that the results suggest the absence of LOX enzymes in rice grains alleviate oxidative deterioration, the molecular determinants which impart the LOX-less characteristics of Daw Dam and CI-115 remain elusive.
Both antisense-mediated and co-suppression-mediated transgenic depletion of genes for LOX have proved useful to elucidate the function of specific LOX enzymes and their corresponding products in plant defense signaling. In Arabidopsis, for example, depletion of a LOX enzyme led to a reduction in the wound-induced accumulation of jasmonic acid (Bell et al., 1995). And results of antisense-mediated depletion of a gene encoding LOX established the involvement of the corresponding enzyme in the incompatibility trait of a tobacco plant resistant to a fungal pathogen (Rancé et al., 1998). A third example where transgenic approaches have been used to elucidate LOX functions relates to the role of a potato LOX, denoted LOX-H1, in growth and development of potato plants (León et al., 2002). It was shown that LOX-H1 depletion resulted in a marked reduction of volatile aliphatic C6 aldehydes, compounds involved in plant defense responses and acting as either signaling molecules for wound-induced gene expression or as antimicrobial substances. A further study showed that transgenic potato plants depleted in the expression of a gene for a LOX enzyme exhibited abnormal tuber development (Kolomiets et al., 2001). However, specific oxylipins that accounted for the tuber phenotype were not identified. In another study, antisense-mediated depletion of potato LOX-H3 suppressed the inducible defense response of the plant, concominant with a higher tuber yield (Royo et al., 1999). Collectively, these data suggest that expression of genes encoding LOX enzymes is important in plant development, possibly with some LOX enzymes playing a defensive role against pathogens, whereas other LOX enzymes generate products that may act to regulate cell development.
It is also of importance to note that tomato fruits with 2-20% reduced levels of two LOX enzymes showed no significant changes in flavor volatiles when compared to wild-type fruits (Griffiths et al., 1999). This finding suggests that either very low levels of LOX are sufficient for the generation of aldehydes and alcohols, or that other LOX enzymes are active in the generation of these compounds.
The role of LOX enzymes is also related to issues outside the field of manufacturing beer, such as LOX-catalyzed generation of hydroperoxy fatty acids that inhibit mycotoxin formation in plants susceptible to fungal contamination, for example as disclosed in U.S. Pat. No. 5,942,661 to Keller. Although the role for LOX enzymes in plant defense and wounding responses remains less clear, the enzymes are induced upon wounding and pathogen challenge (Bell and Mullet, 1991; Bell and Mullet, 1993; Melan et al., 1993; Sarvitz and Siedow, 1996). LOX enzymes' role in wounding and plant defence could be to produce reactive fatty acid hydroperoxides against pathogens (Rogers et al., 1988). Alternatively, LOXs may be induced by stresses to produce signal molecules, such as methyl jasmonate (Bell et al., supra).
Strategies have also been described where 13-HPODE, produced by the action of a LOX enzyme, acts as a substrate for hydroperoxide-converting enzymes to produce flavor-active aldehydes (Noordermeer et al., 2002; Husson and Belin, 2002). Similar processes are disclosed in numerous patents, e.g. U.S. Pat. No. 6,150,145 to Häusler et al. and U.S. Pat. No. 6,274,358 to Holtz et al.
Also, LOX enzymes have been shown to contribute several beneficial effects to bread-making (Casey, 1997). Moreover, U.S. Pat. No. 6,355,862 B1 to Handa and Kausch discloses that fruit quality can be enhanced by inhibiting production of LOX, such as giving a longer shelf life to the product.